Diagnostic accuracy of telestroke consultation: a Louisiana based tele-network experience

BACKGROUND AND PURPOSE: Telestroke has grown significantly since its implementation. Despite growing utilization, there is a paucity of data regarding the diagnostic accuracy of telestroke to distinguish between stroke and its mimics. We aimed to evaluate diagnostic accuracy of telestroke consultations and explore the characteristics of misdiagnosed patients with a focus on stroke mimics. METHODS: We conducted a retrospective study of all the consultations in our Ochsner Health’s TeleStroke program seen between April 2015 and April 2016. Consultations were classified into one of three diagnostic categories: stroke/transient ischemic attack, mimic, and uncertain. Initial telestroke diagnosis was compared with the final diagnosis post review of all emergency department and hospital data. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (LR+) and negative likelihood ratio (LR−) for diagnosis of stroke/TIA versus mimic were calculated. Area under receiver-operating characteristic curve (AUC) analysis to predict true stroke was performed. Bivariate analysis based on the diagnostic categories examined association with sex, age, NIHSS, stroke risk factors, tPA given, bleeding after tPA, symptom onset to last known normal, symptom onset to consult, timing in the day, and consult duration. Logistic regression was performed as indicated by bivariate analysis. RESULTS: Eight hundred and seventy-four telestroke evaluations were included in our analysis. Accurate diagnosis through teleneurological consultation was seen in 85% of which 532 were strokes (true positives) and 170 were mimics (true negatives). Sensitivity, specificity, PPV, NPV were 97.8, 82.5, 93.7 and 93.4%, respectively. LR+ and LR− were 5.6 and 0.03. AUC (95% CI) was 0.9016 (0.8749–0.9283). Stroke mimics were more common with younger age and female gender and in those with less vascular risk factors. LR revealed OR (95% CI) of misdiagnosis for female gender of 1.9 (1.3–2.9). Lower age and lower NIHSS score were other predictors of misdiagnosis. CONCLUSION: We report high diagnostic accuracy of the Ochsner Telestroke Program in discriminating stroke/TIA and stroke mimics, with slight tendency towards over diagnosis of stroke. Female gender, younger age and lower NIHSS score were associated with misdiagnosis.